• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

better: warm water-highflow?cold water low flow

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Hawker-rider

Member
Joined
Apr 13, 2003
Location
AR
Hi guys,

I was thinking here,, what do you think is better, a system with a pump that carries 200GPh and watertemp of 20C or system with 350 GPH and water being 27C??

The whole reason being is that I have a big radiator. and the radiator is most effective with lower flow through it, so it can transfer the most heat. I was wondering if the system will stay colder if the lower flow would actually be better.

Any input?

M
 

Prandtl

Member
Joined
Jan 9, 2003
Location
Montreal, Canada
Hawker-rider said:
the radiator is most effective with lower flow through it, so it can transfer the most heat.

this is not true! I suggest you do some reading, there is some good article in the tips/reviews section of the site.
 
OP
Hawker-rider

Hawker-rider

Member
Joined
Apr 13, 2003
Location
AR
Ok, I did some reading as you suggested, and I found a few irregularities in a few posts that even put physics-equasions into the question. The very interesting thing is here that I didn't see anybody answer is that with all this, there is absolutely no effect of the radiator, which I guess is ok, but that would mean that the radiator has absolutely no effect on heat transfer...... mmmmmm
Q=m*c*dT c being the specific heat of water

Ok what about the following variables that obviously have an effect in heat transfer Surface of heat exchanger both in CPU coller and radiator, cooling effect of the tubing, to make it specific for computer watercooling......

I did find in my setup that if I keep the watercolder and then get my system running the way I want it it stays pretty cool. Once I turn the pump way down and the fans off, I see the watertemp go up..... then when I turn the pump full blast and the fans at full speed I see the water NOT coming back to the same temps as when I start with colder water. I tried with different fanspeeds and pumpspeeds, just seem to come to the same conclusion, when i keep my water colder, my cpu temp will stay a lot lower too.

Anyway I just wanted some input on the better one of the both but the colder water seems to be keeping my cpu colder too.
 

gizxmo

Registered
Joined
May 3, 2003
Location
a
well your flow is a little too much i belive that the system is having problems dissapating all of the heat through the radiator

*the real problem the pump+cpu is too much heat for the radiator with the fan to dissapate either get more radiators with more fans or get a bigger fan*

maybe try buying a couple of heater cores this would exponentually increase the surface area of the dissapation and it would also slow the flow down a tad
 

JFettig

Hey! I showered! Senior
Joined
Jan 5, 2002
Location
MN
true, actually the rad will transfer more heat when the water is going slow, but so will the waterblock. thatll mean the cpu will be hotter, the water temp diff from the inlet to the outlet will be greater, and all that jazz.

where do you get these figures? the temps? they are very unlikely unless you have some really inefficient pumps.

Jon
 
OP
Hawker-rider

Hawker-rider

Member
Joined
Apr 13, 2003
Location
AR
The figures I gave are actually just fictatious ones. For my own watercooling system I have about 240GpH flow with 1/2 ID tubing and a big (approx 13x7inches) heatercore, the heatercore is one of the better ones and I use a windtunnel effect from my fans through my radiator...... yeah... my specialty is aerodynamics in aviation and therefore i did this thing with my watercooling-box ( my setup is outside of my pc-case) So that i have a venturi-pulling air effect over the radiator. The speed of the air is pretty high, especially with the fans running at a higher RPM and i am getting some weird temperature readings, which i also have a few more posts about. At Idle I get about 1-2C BELOW ambient, which some people didn't believe and questioned me about so I decided to measure the temperature of the air going into the heatercore, which ended up being 5-7 BELOW ambient. This is without any refrigeration, mind you.
In Airplanes it is possible for ICE to form out of visible moisture in the air once the temperature drops in the venturi in the carburator, but you have the added effect of fluid added and vaporizing which also withdraws energy from the air. I think i did underestimate the effect of air temperature drop once the speed increases(off course in a converging pipe).

It just was a little odd to me when i was playing around with the nicest point in my setup ( I have a rheostat on both my fans and my pump) that with the pump really low and no fans my water heated up, but after turning everything back to full blast it wouldn't go back to the original values.

I'm 100% positive now that there is a sweet spot in every system that you have to find to get the max out of it. And that's why I was wondering about this.

P.S> mind you, that I am not getting a good overclock in my system, because of my memory, so I'm running close to stock, with only a higher multiplier so my temps shouldn't be that high YET!

Edited ; finished the post finally

M
 
Last edited:

Prandtl

Member
Joined
Jan 9, 2003
Location
Montreal, Canada
JFettig said:
actually the rad will transfer more heat when the water is going slow, but so will the waterblock.

NO, this is not true! Your cpu (when at full load and fix clock speed) produce a fix amount of heat that need to be dissipated.
 

Toysrme

Member
Joined
Nov 12, 2001
Yes! The longer the water stays in contact with the block and pump the more heat it will ahve to absorb.
The longer the water has in the radiator to loose heat, the lower it will be.

The CPU dose have a fixed amount of heat at <insert laod here>
Your problem is that the amount of heat it produces is enough to heat water past it's boiling point. The longer the water is in contact with the heat (anything in the block or pump) the more the temp of the water will rise.

Look... Everything involved with heating in cooling is all about equalibriums<sweet jesus I spelled that correctly!>... Everything is constantly trying to balance itself out. Which is exactly why I love to shoot down the "I'm gonna put TEC's in my res and it'll work" posts. (I always get a chuckle blasting those!)

If you don't belive that, tun off your pump, any modern CPU will fry to death and more in a few min tops. The water will boil and by that time the chip is dead.

This is where the paradox comes in.
If the water is moving slowly so that it's cold when it hits the CPU, remember that hot water is staying in the water block longer. ;) Just a tidbit to ponder.

Everythings a tradeoff ;)